Summary: China’s Ministry of Industry and Information Technology (MIIT) has launched the Industrial Data Foundation Action and pilot work for high‑quality industry datasets, aiming to build consortiums, trusted platforms, and tradable datasets by 2026.
China’s Ministry of Industry and Information Technology (MIIT) issued a notice on March 10 launching the Industrial Data Foundation Action and pilot work to build high‑quality industry datasets that can empower AI. The announcement, reported by Xinhua and echoed in a provincial MIIT forwarding notice, sets a national timeline for data‑sharing consortiums, trusted interconnection platforms, and standardized, tradable datasets to support industry large models and industrial agents by the end of 2026.
What MIIT announced
- Policy launch date: MIIT issued the notice on March 10, formally starting the action plan and pilot work for high‑quality AI‑ready datasets.
- 2026 outcomes: By end‑2026, MIIT expects data collaboration consortiums and trusted interconnection platforms, plus standards and core technologies, to be in place.
- Industrial focus: The pilot program targets key manufacturing industries, pushing AI adoption on the factory floor.
The “1+4+N” structure highlights execution
The notice outlines a “1+4+N” construction approach that connects policy goals to concrete infrastructure. It names four repositories and a broader application layer:
- Industry data resource repository
- Industrial data technology R&D repository
- Industrial data standards repository
- High‑quality industry dataset repository
- N application scenarios to bring industrial large models and industrial agents into real use cases
This structure matters because it moves the story from policy signaling to a blueprint for data infrastructure, standards, and deployment pathways that can be audited and implemented by sector.
Why this matters for industrial AI
Much of the AI conversation still focuses on model releases, but MIIT’s notice treats data readiness as the primary bottleneck. The Xinhua coverage stresses the goal of breaking through data “collection, aggregation, and usage” constraints, which mirrors the shift toward enterprise data platforms such as Huawei’s AI data platform.
If execution follows the stated design, China would move toward a national data‑infrastructure layer that makes industrial AI deployment more repeatable across sectors. That lines up with how policy is intended to reach the factory floor, as seen in Jiangsu’s AI push.
What to watch next
- Consortium formation and data‑sharing governance: Cross‑industry data sharing is hard to coordinate, and early consortium performance will set the pace.
- Regional implementation variance: Funding, governance rules, and sector priorities may vary by province, leading to uneven results.
The action plan should also be read alongside the broader national rollout goals in China’s 2026 AI push for phones, PCs, and robots, which frames industrial AI as one part of a larger policy stack.
More coverage in AI Signals.
